DAB103-Project-Phase01

#%pip install dataprep

#%pip install pandas
#%pip install numpy
#%pip install seaborn
#%pip install matplotlib
#%pip install researchpy
#%pip install scipy
#%pip install statsmodels
#%pip install dataprep
%pip install missingno

# importing necessary libraries
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
%matplotlib inline
import researchpy as rp
import scipy.stats as stats
import statsmodels.api as sm
from statsmodels.formula.api import ols
from dataprep.eda import *
import plotly.express as px
import missingno as msno

#importing the dataset
direct_marketing = pd.read_csv('https://query.data.world/s/lqrsaugj7kwkyazkvydowjuxjrybxx', sep=';')

# New_Name a dictionary
# key = old name
# value = new name
dict = {'age': 'Age',
        'job': 'Job',
        'marital': 'Marital',
        'education': 'Education',
        'default' : 'Credit_default',
        'housing': 'Housing_loan',
        'loan' : 'Personal_loan',
        'contact' : 'Call_type',
        'month' : 'Last_month',
        'day_of_week': 'Last_weekday',
        'duration': 'LastCall_Dur',
        'campaign' : 'NewCampaign_CallNo',
        'pdays' : 'Campaign_Intervals_Day',
        'previous': 'PrevCampaign_CallNo',
        'poutcome' : 'PrevCampaign_result',
        'y' : 'Campaign_Success'      
       }

# call rename () method
direct_marketing.rename(columns=dict, inplace=True)

# print Data frame
direct_marketing.drop(columns= ['emp.var.rate','cons.price.idx', 'cons.conf.idx', 'euribor3m', 'nr.employed'])
display(direct_marketing)

	Age	Job	Marital	Education	Credit_default	Housing_loan	Personal_loan	Call_type	Last_month	Last_weekday	...	NewCampaign_CallNo	Campaign_Intervals_Day	PrevCampaign_CallNo	PrevCampaign_result	emp.var.rate	cons.price.idx	cons.conf.idx	euribor3m	nr.employed	Campaign_Success
0	56	housemaid	married	basic.4y	no	no	no	telephone	may	mon	...	1	999	0	nonexistent	1.1	93.994	-36.4	4.857	5191.0	no
1	57	services	married	high.school	unknown	no	no	telephone	may	mon	...	1	999	0	nonexistent	1.1	93.994	-36.4	4.857	5191.0	no
2	37	services	married	high.school	no	yes	no	telephone	may	mon	...	1	999	0	nonexistent	1.1	93.994	-36.4	4.857	5191.0	no
3	40	admin.	married	basic.6y	no	no	no	telephone	may	mon	...	1	999	0	nonexistent	1.1	93.994	-36.4	4.857	5191.0	no
4	56	services	married	high.school	no	no	yes	telephone	may	mon	...	1	999	0	nonexistent	1.1	93.994	-36.4	4.857	5191.0	no
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
41183	73	retired	married	professional.course	no	yes	no	cellular	nov	fri	...	1	999	0	nonexistent	-1.1	94.767	-50.8	1.028	4963.6	yes
41184	46	blue-collar	married	professional.course	no	no	no	cellular	nov	fri	...	1	999	0	nonexistent	-1.1	94.767	-50.8	1.028	4963.6	no
41185	56	retired	married	university.degree	no	yes	no	cellular	nov	fri	...	2	999	0	nonexistent	-1.1	94.767	-50.8	1.028	4963.6	no
41186	44	technician	married	professional.course	no	no	no	cellular	nov	fri	...	1	999	0	nonexistent	-1.1	94.767	-50.8	1.028	4963.6	yes
41187	74	retired	married	professional.course	no	yes	no	cellular	nov	fri	...	3	999	1	failure	-1.1	94.767	-50.8	1.028	4963.6	no

41188 rows × 21 columns

#exploring the dataset
type (direct_marketing)
direct_marketing.shape
direct_marketing.head()
direct_marketing.tail()
direct_marketing.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 41188 entries, 0 to 41187
Data columns (total 21 columns):
 #   Column                  Non-Null Count  Dtype  
---  ------                  --------------  -----  
 0   Age                     41188 non-null  int64  
 1   Job                     41188 non-null  object 
 2   Marital                 41188 non-null  object 
 3   Education               41188 non-null  object 
 4   Credit_default          41188 non-null  object 
 5   Housing_loan            41188 non-null  object 
 6   Personal_loan           41188 non-null  object 
 7   Call_type               41188 non-null  object 
 8   Last_month              41188 non-null  object 
 9   Last_weekday            41188 non-null  object 
 10  LastCall_Dur            41188 non-null  int64  
 11  NewCampaign_CallNo      41188 non-null  int64  
 12  Campaign_Intervals_Day  41188 non-null  int64  
 13  PrevCampaign_CallNo     41188 non-null  int64  
 14  PrevCampaign_result     41188 non-null  object 
 15  emp.var.rate            41188 non-null  float64
 16  cons.price.idx          41188 non-null  float64
 17  cons.conf.idx           41188 non-null  float64
 18  euribor3m               41188 non-null  float64
 19  nr.employed             41188 non-null  float64
 20  Campaign_Success        41188 non-null  object 
dtypes: float64(5), int64(5), object(11)
memory usage: 6.6+ MB

#exploring possible missing values 
#msno.bar(direct_marketing, figsize=(10, 5), fontsize=10, color='#FFC427', labels = True)
msno.matrix(direct_marketing, labels = True, fontsize = 10, figsize = (10, 5))
plt.title('Number of Missed and Non-missed values in the Datasat')
plt.ylabel('Observations')
plt.xlabel('Variables')

Text(0.5, 0, 'Variables')

plotting the variables starting with numerical variables using px (plotly_express)

fig1 = px.histogram(direct_marketing["Age"], nbins=24, title='Age Distribution',
                   color_discrete_sequence=['#E31737']) #color is based on the theme of the presentation file
fig1.update_layout(xaxis_title_text='Age(years)',
    bargap=0.05, # gap between bars of adjacent location coordinates
    showlegend= False)
fig1.show()

fig2 = px.histogram(direct_marketing['LastCall_Dur'], 
                    nbins=100, title='Last Call Duration Distribution',
                   color_discrete_sequence=['#E31737'])
fig2.update_layout(xaxis_title_text='Last Call Duration (seconds)',
    bargap=0.05, # gap between bars of adjacent location coordinates
    showlegend= False)
fig2.show()

fig3 = px.histogram(direct_marketing['NewCampaign_CallNo'], 
                    nbins=60, title='Distribution of the times that a client has been contacted',
                   color_discrete_sequence=['#E31737'])
fig3.update_layout(xaxis_title_text='Frequency of client contact (times)',
    bargap=0.05, # gap between bars of adjacent location coordinates
    showlegend= False)
fig3.show()

fig4 = px.histogram(direct_marketing['Campaign_Intervals_Day'], 
                    nbins=60, title='Distribution of the days between the last two campaigns contact with clients',
                   color_discrete_sequence=['#E31737'])
fig4.update_layout(xaxis_title_text='Interval (days)',
    bargap=0.05, # gap between bars of adjacent location coordinates
    showlegend= False)
fig4.show()

fig5 = px.histogram(direct_marketing['PrevCampaign_CallNo'], 
                    nbins=60, title='Distribution of the number of the client calls during the last campign',
                   color_discrete_sequence=['#E31737'])
fig5.update_layout(xaxis_title_text='Number of times',
    bargap=0.05, # gap between bars of adjacent location coordinates
    showlegend= False)
fig5.show()

now starting to plot the categorical variables using plotly_express barchart

dm_job = pd.DataFrame(direct_marketing.groupby(['Job']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_job = dm_job.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

fig6 = px.bar(dm_job, x='Job', y='Count',
              color_discrete_sequence=['#009999'])
fig6.update_layout(xaxis_title_text='Jobs',
                   yaxis_title_text='Count',
                  title_text = 'Clients Jobs')
fig6.show()

dm_Marital = pd.DataFrame(direct_marketing.groupby(['Marital']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_Marital = dm_Marital.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

fig7 = px.bar(dm_Marital, x='Marital', y='Count',
              color_discrete_sequence=['#009999'])
fig7.update_layout(xaxis_title_text='Marital statuses of clients',
                   yaxis_title_text='Count',
                  title_text = 'Marital Status')
fig7.show()

dm_Education = pd.DataFrame(direct_marketing.groupby(['Education']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_Education = dm_Education.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

fig8 = px.bar(dm_Education, x='Education', y='Count',
              color_discrete_sequence=['#009999'])
fig8.update_layout(xaxis_title_text='Education statuses of clients',
                   yaxis_title_text='Count',
                  title_text = 'Education Status')
fig8.show()

dm_Credit = pd.DataFrame(direct_marketing.groupby(['Credit_default']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_Credit = dm_Credit.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

fig9 = px.bar(dm_Credit, x='Credit_default', y='Count',
              color_discrete_sequence=['#009999'])
fig9.update_layout(xaxis_title_text='Default credit history of clients',
                   yaxis_title_text='Count',
                  title_text = 'Default credit')
fig9.show()

dm_Housing_loan = pd.DataFrame(direct_marketing.groupby(['Housing_loan']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_Housing_loan = dm_Housing_loan.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

fig10 = px.bar(dm_Housing_loan, x='Housing_loan', y='Count',
              color_discrete_sequence=['#009999'])
fig10.update_layout(xaxis_title_text='Active house loan status of clients',
                   yaxis_title_text='Count',
                  title_text = 'Active house loan status')
fig10.show()

dm_Personal_loan = pd.DataFrame(direct_marketing.groupby(['Personal_loan']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_Personal_loan = dm_Personal_loan.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

fig11 = px.bar(dm_Personal_loan, x='Personal_loan', y='Count',
              color_discrete_sequence=['#009999'])
fig11.update_layout(xaxis_title_text='Active personal loan status of clients',
                   yaxis_title_text='Count',
                  title_text = 'Active personal loan status')
fig11.show()

dm_Call_type = pd.DataFrame(direct_marketing.groupby(['Call_type']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_Call_type = dm_Call_type.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

fig12 = px.bar(dm_Call_type, x='Call_type', y='Count',
              color_discrete_sequence=['#009999'])
fig12.update_layout(xaxis_title_text='Clients device type',
                   yaxis_title_text='Count',
                  title_text = 'Clients device type')
fig12.show()

dm_Last_month = pd.DataFrame(direct_marketing.groupby(['Last_month']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_Last_month = dm_Last_month.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

fig13 = px.bar(dm_Last_month, x='Last_month', y='Count',
              color_discrete_sequence=['#009999'])
fig13.update_layout(xaxis_title_text='Months of the last call of the Client',
                   yaxis_title_text='Count',
                  title_text = 'Last month')
fig13.show()

dm_Last_weekday = pd.DataFrame(direct_marketing.groupby(['Last_weekday']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_Last_weekday = dm_Last_weekday.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

fig14 = px.bar(dm_Last_weekday, x='Last_weekday', y='Count',
              color_discrete_sequence=['#009999'])
fig14.update_layout(xaxis_title_text='Week day of the last call of the Client',
                   yaxis_title_text='Count',
                  title_text = 'Last weekday')
fig14.show()

dm_PrCampaign_Result = pd.DataFrame(direct_marketing.groupby(['PrevCampaign_result']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_PrCampaign_Result = dm_PrCampaign_Result.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

fig15 = px.bar(dm_PrCampaign_Result, x='PrevCampaign_result', y='Count',
              color_discrete_sequence=['#009999'])
fig15.update_layout(xaxis_title_text='Previous campaign result of the contact with clients',
                   yaxis_title_text='Count',
                  title_text = 'The result of the previous campaign')
fig15.show()

dm_Campaign_Success = pd.DataFrame(direct_marketing.groupby(['Campaign_Success']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_Campaign_Success = dm_Campaign_Success.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

fig16 = px.bar(dm_Campaign_Success, x='Campaign_Success', y='Count',
              color_discrete_sequence=['#009999'])
fig16.update_layout(xaxis_title_text='The success for this campaign',
                   yaxis_title_text='Count',
                  title_text = 'The success status for the campaign')
fig16.show()

DAB103-Project-Phase02

#calculating percentage of yes and no in 'Campaign_Success' variable in marital segments 
dm_marital_perc = pd.DataFrame(direct_marketing.groupby(['Marital','Campaign_Success']).size())
dm_marital_perc = 100 * dm_marital_perc/dm_marital_perc.groupby(level=0).sum()
dm_marital_perc = dm_marital_perc.rename(columns={0:'Percent'}).reset_index() 
#if reset_index() is not used, then 'Marital' becomes index and plot can not be drawn
dm_marital_perc = pd.DataFrame(dm_marital_perc)
dm_marital_perc

	Marital	Campaign_Success	Percent
0	divorced	no	89.679098
1	divorced	yes	10.320902
2	married	no	89.842747
3	married	yes	10.157253
4	single	no	85.995851
5	single	yes	14.004149
6	unknown	no	85.000000
7	unknown	yes	15.000000

#testing sns.catplot for next plots
sns.catplot(data=dm_marital_perc,kind='bar',x='Marital' ,hue='Campaign_Success', y = 'Percent').set_titles("{col_name}")

<seaborn.axisgrid.FacetGrid at 0x217ae33a520>

# assigning an age group to each person based on his/her age
age_ranges = [15,30,45,65,100]
age_labels = ['Young','Adult','Mid-aged','Senior']
direct_marketing['Age_Group'] = pd.cut(direct_marketing['Age'], bins=age_ranges, labels=age_labels, right=False)

# adding all categorical variables into a list except the result variable, Campaign_Success
cat_var = ["Job", "Education", "Marital", "Credit_default", "Housing_loan", "Personal_loan",
            "Last_month", "Last_weekday", "Call_type" , "PrevCampaign_result", "Age_Group" ]
# the average success rate among targeted clients in this campaign was 11.625%
average_success_rate = 11.625
# plots colors according to the presentation palette
plot_colors = ['#E31737','#009999']
# repeating the percentage  plot for all categorical variables of the data frame
for i in cat_var:
    dm_perc = pd.DataFrame(direct_marketing.groupby([cat_var[cat_var.index(i)],'Campaign_Success']).size())
    dm_perc = 100 * dm_perc/dm_perc.groupby(level=0).sum()
    dm_perc = dm_perc.rename(columns={0:'Percent'}).reset_index()
    dm_perc = pd.DataFrame(dm_perc) #transforming dataframe to pecentage of campaign_success rate in each category
    sns.set_palette(sns.color_palette(plot_colors)) #using theme colors in plots
    g = sns.catplot(data=dm_perc,kind='bar',x=cat_var[cat_var.index(i)] ,
                    hue='Campaign_Success',y = 'Percent',
                   legend_out = False)
    g.map(plt.axhline, y=average_success_rate, lw = 2.5, ls='--', c='#8497B0') #plotting the horizontal line of overall success rate
    plt.title("Success rate in campaign according to {} status".format(cat_var[cat_var.index(i)]))
    plt.xlabel("{} status".format(cat_var[cat_var.index(i)]))
    plt.ylabel('Percent')
    plt.legend(bbox_to_anchor = (1.01,0.7), title = 'Campaign Success')
    plt.xticks(rotation=45)
    

#this figure is plotted after creating age groups based on age variable
dm_Age_Group = pd.DataFrame(direct_marketing.groupby(['Age_Group']).size()).reset_index() 
#creating a dataframe consisting of the variable and its count
dm_Age_Group = dm_Age_Group.rename(columns={0:'Count'}).sort_values('Count') 
#renaming the column to Count and sorting them

#this figure is plotted after creating age groups based on age variable
fig17 = px.bar(dm_Age_Group, x='Age_Group', y='Count',
              color_discrete_sequence=['#009999'])
fig17.update_layout(xaxis_title_text='Age Group',
                   yaxis_title_text='Count',
                  title_text = 'The age group status for the campaign')
fig17.show()

direct_marketing.head()

	Age	Job	Marital	Education	Credit_default	Housing_loan	Personal_loan	Call_type	Last_month	Last_weekday	...	Campaign_Intervals_Day	PrevCampaign_CallNo	PrevCampaign_result	emp.var.rate	cons.price.idx	cons.conf.idx	euribor3m	nr.employed	Campaign_Success	Age_Group
0	56	housemaid	married	basic.4y	no	no	no	telephone	may	mon	...	999	0	nonexistent	1.1	93.994	-36.4	4.857	5191.0	no	Mid-aged
1	57	services	married	high.school	unknown	no	no	telephone	may	mon	...	999	0	nonexistent	1.1	93.994	-36.4	4.857	5191.0	no	Mid-aged
2	37	services	married	high.school	no	yes	no	telephone	may	mon	...	999	0	nonexistent	1.1	93.994	-36.4	4.857	5191.0	no	Adult
3	40	admin.	married	basic.6y	no	no	no	telephone	may	mon	...	999	0	nonexistent	1.1	93.994	-36.4	4.857	5191.0	no	Adult
4	56	services	married	high.school	no	no	yes	telephone	may	mon	...	999	0	nonexistent	1.1	93.994	-36.4	4.857	5191.0	no	Mid-aged

5 rows × 22 columns

# exploring success rates in different categories of the categorical variables

sns.countplot(x = "Age_Group", hue = "Campaign_Success", data = direct_marketing)

plt.show()

sns.catplot(x="Marital", hue="Campaign_Success", col="Age_Group",
                data=direct_marketing, kind="count")

<seaborn.axisgrid.FacetGrid at 0x217b0c1c640>

# repeating the percentage  plot for all categorical variables of the data frame
for i , j in enumerate(cat_var):
    for k , l in enumerate(cat_var) :
        if i == k:
            pass
        else:
            dm_perc = pd.DataFrame(direct_marketing.groupby([cat_var[k], cat_var[i],'Campaign_Success']).size())
            dm_perc = 100 * (dm_perc/dm_perc.groupby(level=1).sum())/(dm_perc/dm_perc.groupby(level=0).sum())
            dm_perc = dm_perc.rename(columns={0:'Percent'}).reset_index()
            dm_perc = pd.DataFrame(dm_perc)
            g = sns.catplot(data=dm_perc,kind='bar',x=cat_var[i] ,
                            hue='Campaign_Success', y = 'Percent', col = cat_var[k])
            g.set_xticklabels(rotation=45)
            g.map(plt.axhline, y=11.265/dm_perc[cat_var[i]].nunique(), ls='--', c='red')

C:\Users\meali\anaconda3\lib\site-packages\seaborn\axisgrid.py:409: RuntimeWarning:

More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`).

#dropping unnecessary variables
direct_marketing = direct_marketing.drop(columns= ['emp.var.rate','cons.price.idx', 'cons.conf.idx', 'euribor3m', 'nr.employed']) 
direct_marketing.head(1)

	Age	Job	Marital	Education	Credit_default	Housing_loan	Personal_loan	Call_type	Last_month	Last_weekday	LastCall_Dur	NewCampaign_CallNo	Campaign_Intervals_Day	PrevCampaign_CallNo	PrevCampaign_result	Campaign_Success	Age_Group
0	56	housemaid	married	basic.4y	no	no	no	telephone	may	mon	261	1	999	0	nonexistent	no	Mid-aged

# creating a list consisting numerical variables
num_var = ['PrevCampaign_CallNo', 'NewCampaign_CallNo', 'LastCall_Dur']   

#plotting the NewCampaign_CallNo variable in a bloxplot
from matplotlib.ticker import MaxNLocator
#setting 1st and 3rd quartiles and calculating IQR
Q1 = direct_marketing['NewCampaign_CallNo'].quantile(0.25)
Q3 = direct_marketing['NewCampaign_CallNo'].quantile(0.75)
IQR = Q3 - Q1
#calculating whiskers
whis_low = (Q1 - 1.5 * IQR)
whis_high = (Q3 + 1.5 * IQR)
#filtering the outliers
direct_marketing_NewCampaign = direct_marketing.loc[direct_marketing['NewCampaign_CallNo']
                                                           > whis_low]
direct_marketing_NewCampaign = direct_marketing_NewCampaign.loc[direct_marketing_NewCampaign['NewCampaign_CallNo']
                                                                < whis_high]
h = sns.boxplot(x='Campaign_Success', y='NewCampaign_CallNo',
            hue='Campaign_Success',
            data=direct_marketing_NewCampaign)
#setting y axis values to be integers using imported MaxNLocator
h.yaxis.set_major_locator(MaxNLocator(integer = True))
#setting titles and legend
plt.title('Result in campaign according to number of calls made to the client')
plt.xlabel('Campaign Success')
plt.ylabel('Number of calls')
plt.legend(bbox_to_anchor = (1.5,0.7), title = 'Campaign Success')   

<matplotlib.legend.Legend at 0x217bd2b12b0>

#new distribution of "NewCampaign_CallNo" variable
fig3_2 = px.histogram(direct_marketing_NewCampaign['NewCampaign_CallNo'], 
                    nbins=10, title='Distribution of the times that a client has been contacted',
                   color_discrete_sequence=['#E31737'])
fig3_2.update_layout(xaxis_title_text='Frequency of client contact (times)',
    bargap=0.05, # gap between bars of adjacent location coordinates
    showlegend= False)
fig3_2.show()

#plotting the LastCall_Dur variable in a bloxplot after data cleaning
from matplotlib.ticker import MaxNLocator
#setting 1st and 3rd quartiles and calculating IQR
Q1 = direct_marketing['LastCall_Dur'].quantile(0.25)
Q3 = direct_marketing['LastCall_Dur'].quantile(0.75)
IQR = Q3 - Q1
#calculating whiskers
whis_low = (Q1 - 1.5 * IQR)
whis_high = (Q3 + 1.5 * IQR)
#filtering the outliers
direct_marketing_LastCall_Dur = direct_marketing.loc[direct_marketing['LastCall_Dur'] > whis_low]
direct_marketing_LastCall_Dur = direct_marketing_LastCall_Dur.loc[direct_marketing_LastCall_Dur['LastCall_Dur'] < whis_high]
h = sns.boxplot(x='Campaign_Success', y='LastCall_Dur',
            hue='Campaign_Success',
            data=direct_marketing_LastCall_Dur)
#setting y axis values to be integers
h.yaxis.set_major_locator(MaxNLocator(integer = True))
#setting titles and legend
plt.title('Result in campaign according to durations of calls made to the client')
plt.xlabel('Campaign Success')
plt.ylabel('Call duration (seconds)')
plt.legend(bbox_to_anchor = (1.5,0.7), title = 'Campaign Success')   

<matplotlib.legend.Legend at 0x217b0f92ca0>

#new distribution of "LastCall_Dur" variable after cleaning
fig2_2 = px.histogram(direct_marketing_LastCall_Dur['LastCall_Dur'], 
                    nbins=100, title='Last Call Duration Distribution',
                   color_discrete_sequence=['#E31737'])
fig2_2.update_layout(xaxis_title_text='Last Call Duration (seconds)',
    bargap=0.05, # gap between bars of adjacent location coordinates
    showlegend= False)
fig2_2.show()

#repeating above plots in a for loop for numerical variables 
#we don't use this in presentation because the y axis label would be less understandable
for i in num_var[1:]:
    Q1 = direct_marketing[i].quantile(0.25)
    Q3 = direct_marketing[i].quantile(0.75)
    IQR = Q3 - Q1
    whis_low = (Q1 - 1.5 * IQR)
    whis_high = (Q3 + 1.5 * IQR)
    dm_num_var = direct_marketing.loc[direct_marketing[i] > whis_low]
    dm_num_var = dm_num_var.loc[dm_num_var[i] < whis_high]
    sns.catplot(data=dm_num_var,kind='box',x='Campaign_Success' ,hue='Campaign_Success',y = i)
    plt.title('Success rate in campaign according to {} status'.format(i))
    plt.xlabel('Campaign Success')
    plt.ylabel(i)
    plt.legend(bbox_to_anchor = (1.5,0.7), title = 'Campaign Success')